信息技术的发展和大数据的积累驱动了数据科学的发展，人类世界进入大数据时代。大数据时代下的医疗活动也在时刻产生数据医疗大数据，这些数据主要来源于以下4个方面：1、以电子健康档案、医学影像信息为主的临床大数据；2、可穿戴设备、或社交媒体采集的健康相关行为产生的大数据；3、临床试验、基础医学实验产生的生物大数据；4、医疗机构、医疗保险机构等产生的运营大数据。大数据往往具有数据量大、数据流高速、数据类型丰富等特点，健康医疗大数据同样具有上述特点。

大数据技术包含较多功能：数据接入、数据分析、数据处理、数据存储、数据共享交换以及数据展现，其中最主要的功能是数据储存和数据分析、处理。从健康医疗大数据的来源可以看出，医疗大数据主要有结构化数据和非结构化数据构成，并且以非结构化数据为主。结构化数据是指易于搜索、汇总、排序和量化的数据，例如生命体征(如心率)或实验室检测结果，非结构化数据是指不能够预先指定结构的数据，如书面叙述、图像、视频或音频；在医学领域则包括病程记录、病理切片和影像学图像等，这对数据储存提出了较高的要求。然而数据仅存储是毫无价值的，数据要进入分析、处理阶段才能产生可利用的信息。

大数据分析目前已经用于大规模的遗传学研究、公共卫生、个性化精准医疗、新药开发等，对大量数据进行分析往往需要自动化过程，以机器学习为核心的人工智能可以基于数据驱动的算法，学习并且模拟人类的行为，可以高效地处理高维度、大样本、共线性的数据，所以，人工智能在医疗诊断和临床决策中的作用越来越受到重视。

  本文就大数据技术和人工智能在医学实践中的具体应用进行了总结。

The development of information technology and the accumulation of big data drive the development of data science, and the human world has entered the era of big data. Medical activities in the era of big data are also generating data medical big data, which mainly comes from the following four aspects: 1. Clinical big data mainly based on electronic health records and medical image information; 2. Big data generated by wearable devices or health-related behaviors collected by social media;3. Biological big data generated by clinical trials and basic medical experiments;4. Operational big data generated by medical institutions and medical insurance institutions. Big data is often characterized by large amount of data, high-speed data flow and rich data types. Health and medical big data also has the above-mentioned characteristics.

Big data technology includes many functions: data access, data analysis, data processing, data storage, data sharing and exchange, and data presentation, among which the most important function is data storage, data analysis and processing.As can be seen from the source of health and medical big data, medical big data is mainly composed of structured data and unstructured data, and unstructured data is the main one. Structured data refers to data that can be easily searched, summarized, sorted, and quantified, such as vital signs (such as heart rate) or laboratory test results. Unstructured data refers to data that cannot be structured in advance, such as written narratives, images, video, or audio. In the medical field, it includes the course of disease records, pathological sections and imaging images, which put forward higher requirements for data storage. However, data storage is worthless, data must be analyzed and processed to produce usable information.

Big data analysis has been used in large-scale genetic studies, public health, personalized precision medical treatment, development of new drugs, to analyze large amounts of data are often required to process automation, machine learning as the core of artificial intelligence can be based on data driven algorithm, study and simulate the behavior of the human, can efficiently handle the high dimension data, large sample, linear.Therefore, the role of artificial intelligence in medical diagnosis and clinical decision making is receiving more and more attention.

This paper summarizes the specific application of big data technology and artificial intelligence in medical practice.